ENG3036 Simulation of Engineering Systems

Question 1;

a)

In the paper titled A hybrid simulation modelling framework for combining system dynamics (SD), discrete event simulation (DES) and agent-based modes which was published in the year 2020 by LE Khan Ngan Nguyen, Susan Howick and Itamar Megiddo (Nguyen, Howick and Megiddo, 2020), the possibility of solving a DES or SD problem using either DES or SD and their compatibility possibility is extensively discussed. The paper’s main objective was to review and consolidate the existing frameworks for combining the two simulation methods that is DES and SD in order to overcome the challenges posed by use of single simulation methods such as SD. ABM compatibility with SD and DES also discussed.

There are several simulation methods and every method of simulation is designed and developed to serve for a specified purpose. Among the several types of simulation methods, this question discusses two methods namely; Discrete Event Simulation (DES) and System Dynamics (SD) simulation. Of the above two methods of simulation i.e., DES and SD, every particular method has its own specified strength, weakness and special capabilities while some features are shared between the two methods. It can be concluded that it is possible for an expert who only studied DES to study an SD problem since SD simulation model is compatible to DES earthmoving models.

DES is a simulation method having its applications used in numerous fields but it is more perfect and mostly used in queuing networks for stochastic simulation. In the sample, there are several entities collected together and which act and interact as a flow inside the system and for maximum and efficient running for the DES simulation model (Jeong, Lee and Woo, 2017). There are updates for the events in the DES system every time it runs through the flows which is composed of several connected activities and resources and queues. These updates usually occur at discrete time points. The computations of DES simulation models include queued events for down time and up-time running and these events elapse once the model runs through the queued events.

One of the most fundamental and basic SD characteristics is that the its system behavior varies with its structure. SD have a loop called causal-effect loop that is primarily used to generate the variables’ co-functions and relationships. This causal-effect loop is the source of an SD’s power in the field of simulation (AMMON et al., 1997). The behavior of the dynamics that is exhibited in the model system, delays, dependencies among the variables and flows interactions in the system are mainly sourced by the causal-effect loop. SD has an advantage and disadvantage at some points of giving out non-linear and complex feedbacks. Causal Loop Diagrams and Stocks-Flows diagram are the only common types of notation forms that exist in an SD. The Causal Loop Diagrams are responsible for the capture and conceptual relationships in the system while Stock-Flows diagram is responsible for movements description across the entities from the beginning to the termination of the process.

In the process of determining whether an individual who studied DES can solve SD problem, the two simulation methods must be compared and scrutinized accurately. In this study, these two simulation methods will be compared in terms of system structure, problem nature and simulation method. A model might have a system which is discrete, continuous or might have both the features of discrete and continuous systems. For SD models, the variables embedded in the system might be endogenous, excluded or exogenous in terms of the classification of variables process but to DES model these variables have no influence in their output.

In most cases, problem affecting the type of simulation is always related to the simulation model objectives. The model framing which is done in terms of mapping helps in the problem solving. Initially it was said that there are 3 types of system boundaries and both DES and SD have their boundaries ranging among the existing 3 types, these are discrete, continuous or a system having both the boundaries. This implicates it is possible for a DES expert to solve an SD problem since the most crucial part of solving a simulation model is framing the boundary in the system.

A research found out that SD simulation model is compatible to DES simulation model and that the SD model exhibited results that are similar to that of the DES earthmoving model during the study. This point hence proves that a DES expert can study a SD problem since these two simulation methods are compatible. But the research also found out that SD was somehow more superior to the DES earthmoving model hence a DES expert to be very smart in order to solve a problem of a compatible simulation model but more complex one.

b)

System Dynamics’ (SD) information feedback is top-down type while that of Agent Based Modelling has down-top type of feedback hence it is impossible for an ABM expert to study SD problem. The types of feedbacks that these two-simulation methods exhibits are entirely different, these types are also incompatible.

Agent Based Modelling (ABM) exhibit a computational approach and feedback of bottom-up type. Emergent and outcomes which are not intuitive are produced when an ABM simulation model is used. This is because the agents in this simulation method are discrete and they exhibit an autonomous interaction within the simulated space (Tobolic, 2018). There are some rules which are clearly defined that control the rate and type of communication and interaction activities in an ABM simulation model, these clearly defined rules also control the behavior of the simulation outcomes or simulation procedures (Isaac, 2011). Contrary to SD, ABM can only be implemented by the use of programming languages, mostly the third-generation languages. Such languages include Java, Python and C language. Apart from the programming languages, there are some specialized toolkits which can also be used. These specialized toolkits include NetLogo, Repast and Swarm.

The most fundamental and basic SD characteristics is that its system behavior varies with its structure. SD have a loop called causal-effect loop that is primarily used to generate the variables’ co-functions and relationships. This causal-effect loop is the source of an SD’s power in the field of simulation. The behavior of the dynamics that is exhibited in the model system, delays, dependencies among the variables and flows interactions in the system are mainly sourced by the causal-effect loop. SD has an advantage and disadvantage at some points of giving out non-linear and complex feedbacks. Causal Loop Diagrams and Stocks-Flows diagram are the only common types of notation forms that exist in an SD (Arndt, 2006). The Causal Loop Diagrams are responsible for the capture and conceptual relationships in the system while Stock-Flows diagram is responsible for movements description across the entities from the beginning to the termination of the process.

From the study of the ABM and SD structures in terms of functionality and structure, it is crystal clear that an ABM expert cannot study an SD problem. ABM feedback is given in a bottom-up approach while that of SD is given in up-bottom approach. ABM is implemented using programming languages and/or special toolkits while SD implementation does not need any programming language. ABM does have clear-cut system boundaries while SD have system boundaries which may be discrete, continuous or have both. SD applies the use of Causal-loop Diagram and Stocks-Flows diagram as its notation types while ABM uses none.

Question 2;

a)

A DES problem can be studied using SD, this is because both the simulation models are compatible and they also exhibit same types of system boundaries (Tobolic, 2018). There are 3 types of system boundaries; discrete, continuous or a system exhibiting both. It was also found out from a study that SD simulation model is more complex than DES simulation model, thus making it even easier to study DES problem using DES.

One of the most fundamental and basic SD characteristics is that the its system behavior varies with its structure. SD have a loop called causal-effect loop that is primarily used to generate the variables’ co-functions and relationships. This causal-effect loop is the source of an SD’s power in the field of simulation. The behavior of the dynamics that is exhibited in the model system, delays, dependencies among the variables and flows interactions in the system are mainly sourced by the causal-effect loop. SD has an advantage and disadvantage at some points of giving out non-linear and complex feedbacks. Causal Loop Diagrams and Stocks-Flows diagram are the only common types of notation forms that exist in an SD. The Causal Loop Diagrams are responsible for the capture and conceptual relationships in the system while Stock-Flows diagram is responsible for movements description across the entities from the beginning to the termination of the process.

DES ion the other hand is a simulation method having its applications used in numerous fields but it is more perfect and mostly used in queuing networks for stochastic simulation. In the DES system sample, there are several entities collected together and which act and interact as a flow inside the system and for maximum and efficient running for the DES simulation model. There are updates for the events in the DES system every time it runs through the flows which is composed of several connected activities and resources and queues. These updates usually occur at discrete time points. The computations of DES simulation models include queued events for down time and up-time running and these events elapse once the model runs through the queued events (Gustafsson, 2000).

During the process determination whether an individual who studied DES can solve SD problem, the two simulation methods must be compared and scrutinized accurately. In this study, these two simulation methods will be compared in terms of system structure, problem nature and simulation method. A model might have a system which is discrete, continuous or might have both the features of discrete and continuous systems. For SD models, the variables embedded in the system might be endogenous, excluded or exogenous in terms of the classification of variables process but to DES model these variables have no influence in their output (System dynamics, 1981).

In most cases, problem affecting the type of simulation is always related to the simulation model objectives. The model framing which is done in terms of mapping helps in the problem solving. Initially it was said that there are 3 types of system boundaries and both DES and SD have their boundaries ranging among the existing 3 types, these are discrete, continuous or a system having both the boundaries. This implicates it is possible for a DES expert to solve an SD problem since the most crucial part of solving a simulation model is framing the boundary in the system.

A research found out that SD simulation model is compatible to DES simulation model and that the SD model exhibited results that are similar to that of the DES earthmoving model during the study. This point hence proves that a DES expert can study a SD problem since these two simulation methods are compatible. But the research also found out that SD was somehow more superior to the DES earthmoving model. Since the SD model is more complex than the DES model, it makes it even easier for an SD to be used in the study of DES problem.

b)

Discrete Event Simulation (DES) problem cannot be studied using Agent Based Modelling (ABM) since the two simulation models operate in an entirely different way. Agent Based Modelling uses programming languages such as C program, Python and Java for their system implementation. They can also use other special types of toolkits such as NetLogo, Swarm and Repast for system implementation while DES uses none of these for implementation. Since there are great differences between DES and ABM in terms of system structure, problem nature solved and objectives and simulation methods, a DES problem cannot thus be studied using AGM model.

DES is a simulation method having its applications used in numerous fields but it is more perfect and mostly used in queuing networks for stochastic simulation. In the DES system sample, there are several entities collected together and which act and interact as a flow inside the system and for maximum and efficient running for the DES simulation model. There are updates for the events in the DES system every time it runs through the flows which is composed of several connected activities and resources and queues. These updates usually occur at discrete time points (Chamoret, Qiu and Domaszewski, 2009). The computations of DES simulation models include queued events for down time and up-time running and these events elapse once the model runs through the queued events.

Agent Based Modelling (ABM) exhibit a computational approach and feedback of bottom-up type (Bersini, 2012). Emergent and outcomes which are not intuitive are produced when an ABM simulation model is used. This is because the agents in this simulation method are discrete and they exhibit an autonomous interaction within the simulated space. There are some rules which are clearly defined that control the rate and type of communication and interaction activities in an ABM simulation model, these clearly defined rules also control the behavior of the simulation outcomes or simulation procedures. Contrary to DES, ABM can only be implemented by the use of programming languages, mostly the third-generation languages. Such languages include Java, Python and C language. Apart from the programming languages, there are some specialized toolkits which can also be used. These specialized toolkits include NetLogo, Repast and Swarm.

It can be concluded that a DES problem cannot be studied using ABM simulation model. This is due to the clear-cut differences in the structure, implementation methods, simulation methods and problem structure between the two methods that is DES and ABM. The feedback approach type is also different since AGM uses bottom-up feedback approach while DES uses top-bottom feedback approach. Furthermore, the two simulation methods are incompatible showing that these structures are completely different. A simulation method can only be used to study another simulation method problem when they are compatible.

Reference list

, J. (1997). High Performance System Dynamics Simulation of the Entire System Tire-Suspension-Steering-Vehicle. Vehicle System Dynamics, 27(5-6), pp.435–455.

Arndt, H. (2006). Enhancing System Thinking in Education Using System Dynamics. SIMULATION, 82(11), pp.795–806.

Bersini, H. (2012). UML for ABM. Journal of Artificial Societies and Social Simulation, 15(1).

Chamoret, D., Qiu, K. and Domaszewski, M. (2009). Optimization of truss structures by a stochastic method. International Journal for Simulation and Multidisciplinary Design Optimization, 3(1), pp.321–325.

Gustafsson, L. (2000). Poisson Simulation—A Method for Generating Stochastic Variations in Continuous System Simulation. SIMULATION, 74(5), pp.264–274.

He, Y.J. and Fu, L.P. (2013). Simulation Model of Cultural Industry Agglomeration Based on SD-System Dynamics Theory. Key Engineering Materials, 584, pp.312–317.

Isaac, A.G. (2011). The ABM Template Models: A Reformulation with Reference Implementations. Journal of Artificial Societies and Social Simulation, 14(2).

Jeong, Y.-K., Lee, P. and Woo, J.H. (2017). Shipyard Block Logistics Simulation Using Process-centric Discrete Event Simulation Method. Journal of Ship Production and Design.

Nguyen, L.K.N., Howick, S. and Megiddo, I. (2020). A hybrid simulation modelling framework for combining system dynamics and agent-based models. [online] pureportal.strath.ac.uk. Available at: https://pureportal.strath.ac.uk/en/publications/a-hybrid-simulation-modelling-framework-for-combining-system-dyna [Accessed 9 May 2021].

System dynamics. (1981). Mathematics and Computers in Simulation, 23(3), p.312.

Tobolic, T.J. (2018). ABM Clinical Protocols: An ABM Success Story. Breastfeeding Medicine, 13(7), pp.516–516.

Are you in dire need of assignment help in the UK? Can’t figure out who can help you whenever you find yourself thinking, “Wouldn’t it be great if I could pay someone to do my assignment?” With Myassignmenthelp.co.uk, you can fulfil your desires without any hassle.

Send us your requirements, and our paper writers will take care of your assignment worries quickly. So now, you don't have to worry about, "Where can I find someone to do my assignment for me in the UK?” Instead, let our experts provide you with the best assignment help in London, Bristol, Manchester, Liverpool and more!